Addiction can be devastating for addicts, their families, and society as a whole; the emotional and economic costs are astounding. Much of this damage occurs because addiction is a disease of chronic relapse where drug-seeking and drug-taking are repeatedly initiated by exposure to stress, the drug itself, and drug- associated cues. When a gustatory stimulus serves as the cue, rats avoid intake of that taste cue following pairing with a drug of abuse such as morphine or cocaine. For decades, this phenomenon was interpreted as a conditioned taste aversion (CTA) because the paradigm paralleled that form of learning where a taste cue was paired with the aversive consequences of a drug. We have since posed another interpretation based on the anticipatory contrast paradigm where rats avoid intake of a similar taste cue when it predicts access to a highly rewarding sucrose solution. Thus, we hypothesized that rats avoid intake of the taste cue because the value of the cue pales in anticipation of the highly rewarding drug of abuse. In the last decade, we have amassed considerable support for this hypothesis. At the same time, however, we also have uncovered evidence for aversion. Published data reveal the following: Rats avoid intake of a taste cue that has been paired with a drug of abuse. Avoidance of the taste cue is associated with an elevation of the stress hormone, corticosterone, and blunting of the accumbens dopamine peak that normally accompanies ingestion of the sweet taste cue. Greater avoidance of the taste cue is correlated with greater cocaine self-administration and with greater drug- seeking following prolonged abstinence. Finally, intraoral delivery of the drug-associated taste cue elicits aversive taste reactivity (TR, i.e., gapes) and more gaping is associated with a shorter latency to initiate responding for drug, greater load up at the start of the session, and faster acquisition of drug self- administration behavior over trials. Given these data, our working hypothesis is the following: (1) While rats initially avoid intake of the taste cue because it pales in comparison to the potent drug of abuse, ultimately the cue is avoided because it elicits the onset of an aversive state (i.e., conditioned withdrawal) and the onset of this state can be indexed by the onset and frequency of aversive TR. (2) The best correction for this aversive withdrawal state is drug. As such, we predict that the onset and frequency of aversive TR early in training will predict the onset of 'compulsion' (e.g., reduced latency to take the first infusion and greater load up). (3) We hypothesize that aversive TR also will predict the predisposition for 'addiction'. (4) Finally, we predict that manipulations known to affect sensitivity to drug will appropriately affect the onset and frequency of cue- induced aversive TR and corresponding indices of compulsion and addiction.